The h Bar

ACuriousMind

7:00 PM

Weierstraß came back from the grave to do differential geometry?

John Rennie

Thanks :-)

0celouvsky

It's not a geometric problem, its an analysis problem. If $f:A\to\Bbb R$ is smooth in $A^\circ$ and all derivatives are uniformly continuous in $A^\circ$, is $f$ the restriction of a smooth function on $U$, where $U\supset A$ is open?

I'm fairly sure it's true, just lost to history

@ACuriousMind ams.org/journals/proc/1964-015-04/S0002-9939-1964-0165392-8/…

Jaime Gallego

@BernardoMeurer Do you know PDO in PHP?

Bernardo Meurer

@JaimeGallego PHP is contagious and horrible and i refuse to touch it

0celouvsky

salty

Jaime Gallego

7:08 PM

Aw c'mon it's not that bad

Bernardo Meurer

Please stop using it

Whatever reason you have to use it there is something better out there

I don't even wish my enemies to have to use PHP

John Rennie

Yes, you should be using .Net

Bernardo Meurer

If PHP was a weapon it would be a hose. You usually put one end on the car exhaust and another through the window

@JohnRennie NO

John Rennie

@BernardoMeurer :-)

Bernardo Meurer

Use nice tools, like Python with Django or Flask, or Ruby with Rails, or whatever

Don't even learn PHP it will corrupt your brian

John Rennie

7:11 PM

But if you're stuck with maintaining a PHP web site you don't get a choice in the matter.

Bernardo Meurer

Who the hell thought a bash-like syntax for web-oriented programming language was a good idea

@JohnRennie You quit

John Rennie

@BernardoMeurer said by one who doesn't have a mortgage to pay

Bernardo Meurer

@JohnRennie Okay, you suicide

John Rennie

I suppose the life insurance would pay off the mortgage ...

Jaime Gallego

@BernardoMeurer Before I forget, you'll have to keep using that FTP account. SSH access is for my account only.

Bernardo Meurer

7:13 PM

@JaimeGallego I don't mind that, I quite like FTP and filezilla

These files aren't secret or anything

John Rennie

I wrote an FTP version of rsync some decades ago ...

Bernardo Meurer

@JohnRennie Let's talk about my project, you are the perfect adviser. C++ this time :)

With any library I want, this isn't for school

John Rennie

OK the math chat seems to have gone quiet so now is a good time :-)

Bernardo Meurer

So, i'm building a tool that doesn't exist for Linux yet. Not one that I could easily find at least.

John Rennie

Ok ...

Bernardo Meurer

7:16 PM

The objective is to be able to sync two music folders while converting the music files to another format. goals are:
1. Do not re-sync files if no changes were made to the original
2. Keep all metadata
3. Maintain file-structure
4. Be multithreaded

AccidentalFourierTransform

foobar does that

Bernardo Meurer

There will be musync, the music-sync CLI tool, and libsync, a library backend that will, later on, support much more than just music

@AccidentalFourierTransform foobar uses the Win32API, is not open source, and does not run natively on Linux

or BSD or MacOS

This will be cross-platform, boost-based

AccidentalFourierTransform

will it support arbitrary precision arithmetics?

Bernardo Meurer

Maybe

It is my project after all

AccidentalFourierTransform

if you include the picture of a penis somewhere I'll buy it

0celouvsky

7:20 PM

@ACuriousMind Fucking nlab has a page on this :P

Bernardo Meurer

@AccidentalFourierTransform It's free

AccidentalFourierTransform

for now

Bernardo Meurer

It will always be free

AccidentalFourierTransform

you capitalist pig

Bernardo Meurer

GPL licensed

GPL 3.0

FREEDOM

@JohnRennie Anyway, do you get the idea?

John Rennie

7:22 PM

File sync programs aren't very hard. The cool thing about rsync is it can only upload the changed parts of the file, but if you're converting on the fly that isn't going to apply.

Bernardo Meurer

rsync is an amazing tool

John Rennie

So it's just a basic file sync program, but with format conversion on the fly.

Bernardo Meurer

and yeah, for music I could try and only copy over metadata once the soundstream was converted and placed

Hmm

That's not a bad idea actually

But it will be a lot of work

John Doe

Hi all

John Rennie

If you use a server back end you won't be able to sync to removable storage like USB sticks.

Bernardo Meurer

7:24 PM

Hm? Why?

John Rennie

You can't run a server on a USB stick.

Bernardo Meurer

Is this a challenge?

John Rennie

The point of foobar is you can put a USB stick in your PC and sync to it, so you can then put the USB stick in your car radio or whatever.

John Doe

Does it make any sense to write a spin state with a phase i.e. $|\psi \rangle = e^{i \phi}| \uparrow \rangle$?

John Rennie

But then foobar can't sync across the Internet ...

Bernardo Meurer

7:26 PM

@JohnRennie I can add that functionality later

John Rennie

I'm not sure it's obvious why you'd want to sync a low quality version of your music fiels across the Internet.

Bernardo Meurer

@JohnRennie Piracy :P

John Rennie

Syncing across the Internet is a good way of backing up or sharing your files, but you'd want to share or back them up at the top quality.

Bernardo Meurer

Most people don't want 20MB FLAC files for their music

John Doe

@JohnRennie Do you see any problem with writing a spin state with a phase?

Bernardo Meurer

7:30 PM

they want 3MB MP3s

John Rennie

@JohnDoe don't know, sorry.

John Doe

No prob

John Rennie

@BernardoMeurer colour me unconvinced.

Bernardo Meurer

@JohnRennie Okay, do you have blue paint and a large brush?

John Rennie

I suspect there is a reason why this app doesn't exist ...

Bernardo Meurer

7:32 PM

@JohnRennie Which is?

John Rennie

No-one wants it.

Bernardo Meurer

I do :)

I need it a lot actually

I use a tool for that called brutha

And it re-syncs every time, meaning it takes hours even if I just changed a single file

and it also sucks

John Rennie

Link? Googling brutha gets me rap bands.

Bernardo Meurer

All the music players, lollypop, rhythmbox, etc only support MP3, which is patent-encumbered and are bloated with stuff I don't even need for syncing music to my portable player

pypi.python.org/pypi/brutha

ACuriousMind

@JohnRennie Like (Ogg)Vorbis, likely a Pratchett allusion

Bernardo Meurer

7:34 PM

@ACuriousMind I use Vorbis mostly

Opus is much better, but has less support

And both are not patented malware

John Rennie

brutha looks a nice app. It shouldn't resend everything when you change a single file. At least not according to the description.

Bernardo Meurer

Yeah, but it does

Every goddamn time

it's also not multithreaded

John Rennie

Then it has a bug

Bernardo Meurer

it tries to be with GNU parallel

but that doesn't support long paths

and is clunky

it also does not copy over folder.jpg

John Rennie

Bernardo Meurer

7:39 PM

and it's not portable because it needs GNU make, parallel or bash

Yep

I picked it because of that

But it hasn't worked for me

John Rennie

It's written in Python by the looks of it.

Bernardo Meurer

It doesn't actually sync

it creates a file that performs the sync

John Rennie

You should be able to figure out why the resync isn't workign properly.

Bernardo Meurer

It creates a script file basically

John Rennie

OK, so you can figure out why the script file isn't being generated properly.

Bernardo Meurer

7:41 PM

I could, but that doesn't fix the issues:

1) It doesn't do the sync, it creates a script. I don't like that
2) The parallel mode uses GNU parallel which isn't portable and breaks on large paths
3) I'd like a compiled program for this, not a python script
4) Resync doesn't work
5) It doesn't sync cover art files
6) Metadata sometimes gets scrambled

AccidentalFourierTransform

7) has a shitty name and no penis pictures

Bernardo Meurer

It does have a shitty name as well

Mostafa

Happy Adolf Hitler's birthday.

Google homepage

Bernardo Meurer

My birthday is on the day he died lol

ACuriousMind

@Mostafa That's a joke, right?

Bernardo Meurer

7:45 PM

@ACuriousMind yep, I just checked

John Rennie

Origin appears to be here. A joke in dubious taste.

Anonymous

Is that even a joke?

John Rennie

@BernardoMeurer I have to get to bed. I'm at work in 8 hours. I'll sleep on the sync project.

Bernardo Meurer

I didn't even laugh about this one, and I have the worst of tastes

@JohnRennie Cool, thanks :) Goodnight!

Anonymous

@BernardoMeurer Nice new avatar :D

Slereah

7:51 PM

Geroch's answer on non-Hausdorff Lorentz structures :

I don't know what the "official" definition is, but here is what I
would regard as a reasonable one:

Let M be a (possibly non-Hausdorff) manifold. By a <tensor field>
on M we mean, for every open, Hausdorff subset U of M, the
specification of such a tensor field on U, such that: Given
any U and U', the two tensor fields agree on their intersection.

I cannot imagine any other definition. Anyway, with this definition
there certainly exist non-Hausdorff manifolds with Lorentz metrics.
For example, take two copies of the region t >= 0 of Minkowski

Bernardo Meurer

@blue Thanks man! Got it on my Easter trip to the Azores

Anonymous

@BernardoMeurer Oh, how was the trip? (Other than the slow internet :P)

Anonymous

I've heard Azores is a famous tourist spot :)

Bernardo Meurer

@blue Wonderful! Most beautiful place I've been to nature-wise. Coming from someone from Rio it's a big statement :P

Mostafa

@BernardoMeurer It seems that it's only visible to users from some areas (probably Achsenmächte countries:/ )

Anonymous

7:54 PM

@BernardoMeurer Ah, good that you enjoyed :)

Mostafa

@ACuriousMind ^

Anonymous

@Mostafa Which country's Google server are you using? I can't find it on the German one...

Bernardo Meurer

@Mostafa I'm pretty sure it's fake

ACuriousMind

@Mostafa It's not true, stop trolling.

Mostafa

OK, source .

ACuriousMind

7:56 PM

JR already linked that :P

Anonymous

@Mostafa That's not even a source....

Anonymous

Not an official one atleast

Bernardo Meurer

It's obviously photoshopped

What year are you people from? 2004? To fall for this bullshit it must be

AccidentalFourierTransform

AdS/CFT is bullshit

ACuriousMind

@AccidentalFourierTransform Hmmmm?

AccidentalFourierTransform

8:02 PM

it so is

Mostafa

@AccidentalFourierTransform So you finally came up with your original idea for your thesis....

AccidentalFourierTransform

oh no, I have plenty of original ideas

the best one is that a photon has an E=hf wave nature, and two waves can be superposed

and that an electron has a spin ½ nature, such that its electromagnetic field is not totally unlike the "vortex" of gravitomagnetism

For a example you can grasp intuitively, think of two whirlpools. Two whirlpools cannot overlap.

Mostafa

8:23 PM

@ACuriousMind What about this one?

I checked 123456 as password for the university emails of all the faculty members of my (previous) university. it worked for 6 emails :|

one of them is from the math department. others from mining engineering, civil engineerig and metallurgy engineering departments.

Ben Niehoff

9:32 PM

lol AdS/CFT is so not bullshit

0celouvsky

9:47 PM

@BenNiehoff what

@ACuriousMind The proof...appears to be in Hörmander? I can't tell, these old PDE books love to be mysterious.

DanielSank

I just found the most incredible example of bad grammar on this site.

Well, punctuation

> I can't even continue with this equation, because it's simply zero? This doesn't make sense to me? Can someone please explain how I obtained zero..

@heather @ACuriousMind

0celouvsky

why not @ me

DanielSank

@Danu

It's just amazing.

@0celouvsky

0celouvsky

too late now!

DanielSank

nope

They got all three termination characters wrong.

They put ? instead of . in the first two sentences.

Then they put .. instead of ? in the final sentence.

ACuriousMind

10:04 PM

Maybe they think punctuation marks are to signal the cadence?

Maybe they're just confused about what punctuation is, though

Btw, I recently learned that the odd habit of some users to put a space before punctuation is actually correct in the "French style" of punctuation typography

Alas, since few of them are French, that's probably not their excuse

0celouvsky

huh

0celouvsky

10:32 PM

@ACuriousMind @Slereah @BalarkaSen "I personally feel that the next person to propose a new definition of a connection should be summarily executed." - Mike Spivak.

Bernardo Meurer

@0celouvsky Tell reb my phone died and I am reviving it

0celouvsky

done

Bernardo Meurer

Also, Michelle talked to me these last couple days, she's alright

and her birthday gift to me arrived :)

A shirt and a book :D

0celouvsky

She's practicing right now

Bernardo Meurer

Ah, alright

Which piece?

Okay, @0celouvsky, my phone is charging and if I move it one centimeter it stops. So tell her "That's alright, I'm pulling an allnighter for Analysis, ping me when you're back and we solve the issue"

0celouvsky

10:38 PM

Nielsen concerto

@BernardoMeurer strange phone

Bernardo Meurer

It's > 5 years old

The Apple lightning connector absolutely sucks

ACuriousMind

...you're using an apple phone? You, the anti-malware prophet?

0celouvsky

screw you ACM

@BernardoMeurer I don't think she knows humbolle

Bernardo Meurer

@ACuriousMind I bought it >5 years ago :)

If you want to buy me an android and help me out, please do

0celouvsky

wtf

apple is the best phone

Bernardo Meurer

10:41 PM

I don't disagree

ACuriousMind

@BernardoMeurer I...might have an old one lying around

Bernardo Meurer

It's just not the best company

Battery life sucks on apple though

0celouvsky

why does everyone in this chat get things except for me

Btw, obe is alive.

Bernardo Meurer

@0celouvsky You have an iPhone 7, an almost brand new top-of-the-line MacBook Pro, and more books than most of us. Shut up.

Oh, that's good news! What happened to him?

0celouvsky

He doesn't think he belongs here

Bernardo Meurer

10:46 PM

Well, that's stupid

0celouvsky

I can assist with analyiss

I've got nothing going on

ACuriousMind

None of us belongs here. We're all lost in this sea of puns, math, food and the occasionally surprising occurence of physics

Bernardo Meurer

@0celouvsky Sweet, I'm in school, let me grab my exercises

Cows

Going for a walk. Will be listening to this youtube.com/watch?v=-xZJq1vdsjQ

0celouvsky

@ACuriousMind I met a PDE guy just now at an awards thing and rambled about Sobolev spaces. He suggested some papers and then admitted that no one bothers to check these things because they're probably correct

I really do not understand that mentality

Bernardo Meurer

10:49 PM

@ACuriousMind Me and JR talked physics for hours straight the other day, and you didn't want to join

ACuriousMind

@0celouvsky Remember that "mathematics, morally", thing? It's another instance of it

@BernardoMeurer I don't recall that - was I even here during that conversation? :P

Bernardo Meurer

@ACuriousMind Yes

And you said it was "morning" but I was awake, so that is impossible

ACuriousMind

lol, I say "morning" to the first few hours after I got up, regardless of actual daytime :P

Bernardo Meurer

math.tecnico.ulisboa.pt/~jpnunes/CII/exames/T1-12-11-16.pdf

@0celouvsky For starters I need to solve this

Okay

I have no clue what I'm doing

0celouvsky

Wait, I need to figure out this sheaf notation

I blame @ACuriousMind

Bernardo Meurer

10:55 PM

ACM is the reason I suck at analysis

0celouvsky

ban @ACuriousMind

Bernardo Meurer

@ACuriousMind Ban yourself

Or help me with analysis

your choice

0celouvsky

I think they mean the sections of the sheaf

Bernardo Meurer

Pretty sure banning yourself is the smart option

0celouvsky

@BernardoMeurer She wants me to listen to her play so idk if I can skype

I have to go someplace far away and don't know if I'm coming back to my room

Bernardo Meurer

10:57 PM

I can't skype either, so it's okay

ACuriousMind

"someplace far away" sounds ominous

0celouvsky

@BernardoMeurer ok

@ACuriousMind you don't look very banned

Bernardo Meurer

'bout time

0celouvsky

@ACuriousMind the music building

ACuriousMind

@0celouvsky I actually checked, I cannot ban myself

0celouvsky

10:58 PM

@ACuriousMind damn!

ACuriousMind

@QPaysTaxes Yes. Not much to see here, carry on

0celouvsky

@BalarkaSen Sure, but I am wondering if Lorentz metric --> Hausdorff.

@ACuriousMind Bernardo apologized to me and asked that his suspension be moved to Friday after 11AM.

I forgive him.

ACuriousMind

@0celouvsky If you didn't want him to get suspended you shouldn't have flagged that. That message actually was rude (although I would have let it slide with just deletion and not suspension had you not flagged it).

0celouvsky

I wanted to get it deleted.

@ACuriousMind You seriously won't lift the ban? I misclicked.

@ACuriousMind He will have to work at Burger King if you don't lift the ban

ACuriousMind

@0celouvsky Okay, here's what we're gonna do: I'm gonna lift the suspension and next time he (or someone else) says something you don't like try talking to them first instead of flagging directly, okay? (Unless it's really egregious, of course. Use your reason)

0celouvsky

11:05 PM

Thank you

Bernardo Meurer

Phew

God bless

0celouvsky

I suggested that policy long ago but it was poopooed

Bernardo Meurer

Thanks @ACuriousMind @0celouvsky

Okay

0celouvsky

@BernardoMeurer PAPA bless you heathen

Bernardo Meurer

What is a limit

0celouvsky

11:05 PM

$x\to y$

Bernardo Meurer

in $\Bbb R^n$

math.tecnico.ulisboa.pt/~jpnunes/CII/exames/T1-12-11-16.pdf

The first function

the limit in (a) is 0, no?

0celouvsky

How did you decide that?

Bernardo Meurer

Well, the function is 0 at (0,0)

so when it goes to (0,0) the limit is 0?

0celouvsky

yikes

uh, no

Bernardo Meurer

:/

How the heck do I do this?

@ArtOfCode Have you ever watched The Office?

ArtOfCode

11:08 PM

can't say I have

0celouvsky

My gut feeling is that the limit doesn't exist

Bernardo Meurer

@0celouvsky How do I verify that?

@ArtOfCode Watch it. You are Toby from that series

ArtOfCode

why do I get the feeling that isn't a compliment

Bernardo Meurer

Because you know me

0celouvsky

actually it is zero

Bernardo Meurer

11:10 PM

How do I decide whether or not it exists?

0celouvsky

$$|\frac{y^3}{x^2+y^2+x^4}|\le |\frac{y^3}{y^2}|=|y|\to 0$$

Ask @ACuriousMind to explain (nicely), I'll brb

Bernardo Meurer

@ACuriousMind Explain nicely

Please

ACuriousMind

I'm thinking. If you are confused about the concept of limit, it makes little sense to look at that particular function

Bernardo Meurer

No, no

I understand a limit in $\Bbb R$

I just want to know how do I know if a limit exists in $\Bbb R^n$

ACuriousMind

@BernardoMeurer Ah, well - basically you just replace all absolute values in $\mathbb{R}$ with the norm $\mathbb{R}^n$.

0celouvsky

11:16 PM

Given epsilon, find a delta

Bernardo Meurer

@ACuriousMind Example?

@0celouvsky No epsilons and deltas, they are worse than sea-sickness and spaghetti

Ah

ACuriousMind

@BernardoMeurer Well, I mean, in $\mathbb{R}$ you say that $x$ is the limit of a sequence $x_n$ if $\lvert x_n - x\rvert$ gets arbitrarily small for large $n$ (I'm paraphrasing the $\epsilon$-$\delta$ definition here). In $\mathbb{R}^n$ you say that $\vec x$ is the limit of $\vec x_n$ if $\lvert\lvert \vec x - \vec x_n\rvert\rvert$ becomes arbitrarily small

Bernardo Meurer

I can verify whether the limits as x approaches 0 and y approaches 0 are the same, no? That's necessary but not sufficient

@ACuriousMind Ah, right! Because the norm is the analogous structure (in $\Bbb R^n$)to a modulus (in $\Bbb R$)

Okay

@ACuriousMind on 1.b) here won't the derivative be $0$ since the function is just constant at that point?

ACuriousMind

@BernardoMeurer What does it mean to be "constant at a point"?

Bernardo Meurer

Nothing

Everything is constant at a point

ha

ACuriousMind

11:24 PM

Exactly ;)

Bernardo Meurer

So no

I have to take the derivative of the other part of the function?

ACuriousMind

@BernardoMeurer Yes - write down the derivative in terms of its definition as a limit of a difference quotient, then figure out whether that limit exists

Bernardo Meurer

$$\lim_{h\to 0}\frac{||f(a+h) - f(a) - Df(a)||}{||h||} = 0$$

This is what I have on my notes

@ACuriousMind Is this what you're looking for?

ACuriousMind

@BernardoMeurer Yes. The art is now to make an educated guess for what $Df(a)$ could be at $a=0$, or to show no such $Df(0)$ can exist

Bernardo Meurer

It's going to be a $2\times 2$ matrix, right?

ACuriousMind

11:32 PM

Well, actually, that question is just asking for the partial derivative in $y$, so what's the definition of that?

Bernardo Meurer

@ACuriousMind The only definition I have is the above, all I know is that in a partial derivative I treat the other variables as constants

ACuriousMind

You don't have a proper definition of what $\partial f / \partial y$ is supposed to be?

Bernardo Meurer

::looks away in shame::

Oh

No, yeah, I don't

I have a bunch of stuff over some vector

ACuriousMind

@BernardoMeurer So, you take the partial derivative as just adding the $h$ to the coordinate you're differentiating with respect to

That is, $\partial f/\partial y (x_0,y_0) = \lim_{h\to 0} \frac{f(x_0,y_0 + h) - f(x_0,y_0)}{h}$

(where $h$ is just a real number)

Bernardo Meurer

Oh, nice, I had just written that too

Why no norms?

ACuriousMind

11:45 PM

they're hidden in the $\lim$ ;)

Bernardo Meurer

@ACuriousMind Ah

Okay, so that limit goes to infinity, no?

ACuriousMind

@BernardoMeurer Why?

Bernardo Meurer

Wait wait

Is it a 0/0 deal?

f(0,0) is just 0

so w/e

now f(0,0+h) as h goes to 0...

Goes to 0

and h goes to 0

Do I really need to use hospital on this ::rolls eyes::

ACuriousMind

@BernardoMeurer Stop! Don't consider numerator and denominator separately

Bernardo Meurer

O.O

ACuriousMind

11:51 PM

What is $f(0,h)/h$?

Since that's the expression you're taking the limit $h\to 0$ of.

Bernardo Meurer

What do you mean by "What is"?

ACuriousMind

@BernardoMeurer When you replace $f(0,h)$ by what it actually is and then simplify that expression, what is the result?

Bernardo Meurer

1!

ACuriousMind

Aha. So...what is $\lim_{h\to 0} 1$?

Bernardo Meurer

73!

ACuriousMind

11:53 PM

Close :P

Bernardo Meurer

1!

ACuriousMind

Right, so there's your answer

Bernardo Meurer

So there is no partial derivative!

At that point

ACuriousMind

@BernardoMeurer Why not?

Bernardo Meurer

Because $1 \neq 0$?

ACuriousMind

11:56 PM

And why should it be $0$?

Bernardo Meurer

Because I'm stupid

One moment

So now I also have to compute the partial derivative

and then see if it's the same at the point

ACuriousMind

@BernardoMeurer Did we not just compute the partial derivative?

Bernardo Meurer

I thought we were checking if it existed!

How do I know it exists?

ACuriousMind

Yes, and in that process we computed it!

Bernardo Meurer

I mean, we can compute it and it not mean anything

So, for it to exist the limit must be in $\Bbb R^n$?

That's all

and the limit will be the partial derivative

ACuriousMind

11:59 PM

If it didn't exist, we would have found that the limit we were trying to take did not exist, but fortunately, $\lim_{h\to 0} 1$ exists

Transcript for